This was a cross-sectional study conducted at 10 US EDs in 2008-2009. The study questionnaire was conducted as the initial assessment of a prospective cohort study of smoking and novel predictors of smoking cessation among ED patients. During a 10-day enrollment period, trained research staff screened consecutive ED patients for tobacco use during peak volume hours (9:00 AM to midnight). Each site enrolled a minimum of 36 subjects.
All subjects received treatment-as-usual by ED providers for their tobacco use; however, research staff gave subjects an educational pamphlet on smoking cessation published by the US Department of Health and Human Services and a list of tobacco-cessation treatment options, which included the National Quitline number. Furthermore, subjects who screened positive for depression, alcohol, or drug use were given educational pamphlets published by the Association for Behavioral and Cognitive Therapies. In addition, subjects received brochures with national mental-health hotlines and state-based behavioral health referral services, which could be used to identify mental-health and substance-abuse treatment options.
The Institutional Review Boards of the participating institutions approved this study.
Study setting and population
The 10 participating EDs (located in Phoenix, AZ; Loma Linda, CA; Denver, CO; Baton Rouge, LA; Camden, NJ; Bronx, NY; New York, NY; Akron, OH; Cleveland, OH; and Portland, OR) were selected for their diverse geographic, socioeconomic, and ethnic/racial patient populations. All were urban teaching-hospital EDs staffed by attending and resident physicians and physician extenders. The annual patient volume at these institutions ranged from 35,000 to 87,000. The majority serve a population that is > 25% African American, and four serve populations that are > 20% Hispanic. More than 20% of patients in most of the participating institutions lacked health insurance. No participating ED had formal tobacco intervention services or treatment protocols that would distinguish them from the typical US ED.
Patient inclusion criteria were age 18 years or older and daily or current occasional cigarette use. We excluded patients with altered mental status, acute intoxication, hostile or agitated behavior, an insurmountable language barrier, transient residence, lack of access to a telephone, or severe illness that would preclude conversation. Sites maintained a registry with all patients registered in the ED during the shift to facilitate a comparison of enrolled patients to those not enrolled.
Subjects completed a self-report, paper-and-pencil baseline assessment during their ED visit. The assessment included questions on smoking-related variables, predictors of cessation, and interest in a variety of hypothetical ED-initiated tobacco interventions and counseling styles. It took 15-20 min to complete. All measures were printed in both English and Spanish. The assessment was completed through research staff interview when necessary to accommodate patients with poor eyesight or illiteracy. This was done for < 5% of subjects. To reduce demand bias, which could lead to under-reporting of tobacco use and over-reporting of interest in cessation or interventions, participants were reassured that their responses would not be revealed to their treating clinicians. The specific variables we assessed are described under the Measures section.
In addition to the self-report measures, trained research staff completed a structured chart review for each subject, noting a variety of characteristics associated with the visit including the diagnoses assigned to the ED visit, triage level, disposition (admitted versus discharged), and documentation of smoking-related management, such as whether the ED clinicians provided smoking cessation counseling or referrals to the individual.
The study was coordinated by the Emergency Medicine Network (EMNet). Data collection forms were reviewed by EMNet staff, and missing or inconsistent data were reconciled through communication with the site. All data underwent double data entry.
A broad range of measures were included in the parent study. The focus of this paper is on the interest in ED-initiated interventions and counseling styles. The specific measures used in the data analyses are described below.
Demographic characteristics included age, gender, race, ethnicity, education, insurance status and income.
Tobacco use and nicotine dependence
Current smokers were defined as anyone who had smoked at least 100 cigarettes in their lifetime and reported currently smoking cigarettes every day or some days. Although most randomized trials of tobacco interventions only include daily smokers, including nondaily smokers helps account for the tendency of medically ill smokers to temporarily decrease or stop smoking because they are feeling poorly, not because they are actively trying to quit [21
]. Nicotine dependence was assessed with the Heavy Smoking Index (HSI) [22
], a well-established two-item self-report measure for use when rapid assessment is needed. Strength of nicotine dependence is represented by the sum of cigarettes smoked per day (0 = 1-10; 1 = 11-20; 2 = 21-30; 3 = 31+) and the time until first cigarette (0 = 61+ minutes; 1 = 31-60; 2 = 6-30; 3 = 0-5). Scores between zero and three indicate low to moderate dependence, and scores greater than three indicate high dependence. The HSI correlates highly with the Fagerström Test for Nicotine Dependence, the most widely use measure of nicotine dependence, and has been shown to be positively associated with carbon monoxide levels [24
Interest in ED-initiated interventions
We created a list of possible interventions that could be initiated in the ED and assessed patient interest using a 5-point Likert-type scale (1 = not at all interested, 5 = extremely interested). The interventions consisted of 1) reading a pamphlet or watching a video that describes ways to stop smoking, 2) receiving a list of telephone numbers for places to get stop-smoking counseling, 3) having one's name and telephone number sent confidentially to a stop-smoking counselor so he/she can call you at home to discuss treatment options, 4) having an actual appointment with a stop-smoking counselor scheduled within the next four weeks (before leaving the ED), 5) stop-smoking counseling during the current ED visit, 6) getting a prescription for NRT, such as the patch, gum, lozenge, or spray, upon discharge, 7) getting a prescription for medication that help people quit, like bupropion or varenicline, at discharge, and 8) being enrolled in an 8-week stop-smoking program. Individuals who expressed interest in ED counseling were asked how many minutes would be acceptable: 0-5 min, 6-15 min, 16-30 min, 30-45 min, or "no limit, as long as it doesn't delay my care."
We asked subjects to rate their acceptance of a variety of different counseling messages on a 5-point Likert type scale (1 = strongly dislike, 5 = strongly like). The counseling messages were 1) the counselor or doctor explains to you the health risks of smoking; 2) the counselor or doctor tells you that you should quit smoking immediately; 3) the counselor or doctor asks you some questions that help you to identify your own reasons for quitting, as well as barriers that prevent you from quitting; 4) the counselor or doctor explains that you are addicted to nicotine; 5) the counselor or doctor asks you some questions that help you identify high-risk situations and teaches you how to avoid smoking in these situations; 6) the counselor or doctor shows you pictures of people's lungs after they have smoked for years to try to scare you into quitting; 7) the counselor or doctor tells you that you should be ashamed of yourself for smoking; 8) the counselor or doctor explains that it is your own choice of when and how you quit; and 9) the counselor or doctor does an assessment and gives you feedback on how smoking has already affected your health. Some items were consistent with motivational interviewing (MI) principles, while others were not.
We assessed a list of smoking-related symptoms and diagnoses over the past 12 months using "yes/no" questions, including heart disease or heart attack, high blood pressure, stroke, problems with blood circulation, peripheral vascular disease, cancer, chronic obstructive pulmonary disease (COPD), emphysema, bronchitis, congestive heart failure, wheezing, shortness of breath, respiratory or sinus infection, cough, congestion, pneumonia, and asthma. A sum score was calculated representing the total number of symptoms and diagnoses endorsed.
All International Statistical Classification of Diseases and Related Health Problems
ed.) (ICD-9) diagnoses assigned by trained ED coders and used for billing purposes were obtained on all subjects. The diagnoses were categorized based on whether they met criteria for a smoking-related disease as outlined by the US Surgeon General. This is a commonly used strategy to classify smoking-related diseases and has been applied successfully to ED patients [25
]. Two variables were created based on whether the primary ED diagnosis was smoking-related and whether any ED diagnosis was smoking-related.
Emergency department evaluation and management
Trained research assistants (RAs) reviewed medical records of participants and abstracted relevant visit data using a standardized form. Charts were reviewed for clinical data, including triage level, patient disposition, and ICD-9 diagnoses assigned to the visit, and for documentation of patient tobacco use. If smoking was documented, RAs noted if clinicians indicated amount of tobacco use (packs per day or years) and if they provided counseling, smoking discharge instructions, NRT, or referrals to outpatient quitting resources.
Descriptive statistics are presented as means with standard deviations or counts with percentages and 95% confidence intervals. For the primary objective of the study (to examine patient preferences regarding tobacco interventions), in addition to reporting the results based on a 5-point Likert scale, interest in each counseling type was converted to a dichotomous variable (interested/not interested). As participants rated preferences on a 5-point scale, with 3 as the "neutral" option, interest in an intervention was defined as a score greater than 3.
For the secondary research objective (to examine how patient characteristics affected preferences for receiving ED-based counseling for smoking cessation), we first examined variables likely to be influential on smoking based on the literature, including demographics (age, gender, race/ethnicity, education), amount of smoking (by HSI), smoking-related symptoms, and smoking-related diagnoses [26
]. We assessed these variables individually for association with interest in ED counseling using t-tests or Pearson correlation as appropriate. Finally, we developed a linear regression model for interest in ED counseling, including variables that reached statistical significance by p
< 0.10 in the univariable analysis. All analyses were performed using SPSS 17.0 (Chicago, IL, USA).